Tracking and people counting using Particle Filter Method

In recent years, technology has developed quite rapidly, especially in the field of object tracking. Moreover, if the object under study is a person and the number of people a lot. The purpose of this research is to apply Particle Filter method for tracking and counting people in certain area. Track...

Full description

Saved in:
Bibliographic Details
Published in:Journal of physics. Conference series 2018-03, Vol.974 (1), p.12059
Main Authors: Sulistyaningrum, D R, Setiyono, B, Rizky, M S
Format: Article
Language:English
Subjects:
Centroids
Model testing
Occlusion
Physics
Probabilistic models
Segmentation
Tracking
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In recent years, technology has developed quite rapidly, especially in the field of object tracking. Moreover, if the object under study is a person and the number of people a lot. The purpose of this research is to apply Particle Filter method for tracking and counting people in certain area. Tracking people will be rather difficult if there are some obstacles, one of which is occlusion. The stages of tracking and people counting scheme in this study include pre-processing, segmentation using Gaussian Mixture Model (GMM), tracking using particle filter, and counting based on centroid. The Particle Filter method uses the estimated motion included in the model used. The test results show that the tracking and people counting can be done well with an average accuracy of 89.33% and 77.33% respectively from six videos test data. In the process of tracking people, the results are good if there is partial occlusion and no occlusion
ISSN:1742-6588
1742-6596
DOI:10.1088/1742-6596/974/1/012059